Estimation of plasma density perturbation from dusty plasma injection by laser irradiation on tungsten target in DiPS

In Je KANG,Min-Keun BAE,In Sun PARK,Min Ji LEE and Kyu-Sun CHUNG

1 Department of Electrical Engineering,Hanyang University,Seoul 133791,Republic of Korea

2 Plasma Technology Research Center,National Fusion Research Institute,Gunsan 54004,Republic of Korea

Abstract

Keywords: plasma density perturbation,dusty plasma,tungsten dust,DiPS

1.Introduction

In the fields of laboratory,astrophysical,and fusion plasmas,studies on dusty plasmas have been carried out on the mechanisms of charged dust particles,dust acoustic waves,dust particle transport,and their effects on plasma characteristics[1–4].The analysis of dusty plasmas requires understanding of the fundamentals of plasma and atomic physics,innovative experiments and diagnostics,environmental issues,and novel industrial applications because dust particles fully interact with and are coupled to the background plasma,resulting in new and unique plasma phenomena to arise in the plasma [3,4].

In fusion devices,the studies have focused on dust particle transport,the interaction of dust with plasmas,the observation of dust parameters,and dust effects on edge and core plasmas in terms of various simulation or experimental conditions[2,4–6].Although some mechanisms leading to the formation,transport,and effects of dust on core plasmas have been analyzed,their relative importance is not yet adequately understood [7].Additionally,significant amounts of dust particles,which are a few μm in size,have been found at the divertor or bottom region,and the negative effects of dust on core plasmas have been shown in many recent studies of advanced fusion devices such as JET [8],ASDEX-U [9],DIII-D [10],and KSTAR [11,12].As the development of fusion devices more closely and reliably approaches the technical conditions required for the confinement of a core plasma,it clearly appears that optimization of the plasma performance requires improved understanding and ultimately better control of the interface conditions for dust-interaction with the plasma and material surfaces.This is due to the fact that the major concern so far has been pollution of the plasma by ablated dust particles,which may decrease the performance of the plasma [2,7,13,14].

Many studies with a laser,an electron beam,and a plasma gun for the interaction of transient heat flux with tungsten materials have been carried out in the research field of plasma wall interactions for fusion devices in recent years.This is because tungsten has been selected as a plasma facing component(PFC) at the divertor baffles and dome in the International Thermonuclear Experimental Reactor (ITER) and proposed as a first wall material in DEMO [15–17].The surface damage of tungsten from transient heat flux such as edge localized modes(ELMs) was investigated when ELM-like conditions were replicated by exposing the tungsten surface to a pulsed laser beam[18,19].Relevant results for the effects of ELMs on tungsten surfaces in fusion devices were presented from the experimental simulation of laser-tungsten interactions,such as melting,cracks,recrystallization and the He bubble structure [16–21].However,the research for the secondary effects on the confined plasmas such as the core or edge region at fusion devices is absent,although dusty plasmas are generated via interactions between the transient heat flux and tungsten,producing dusts from the volume difference according to the vaporization,erosion,melting,and cracks of tungsten surfaces [19,21].

The divertor plasma simulator (DiPS) was developed as a linear plasma device to apply the probe technique to understand a magnetized presheath region of a simulated tokamak plasma wall interaction region [22,23].DiPS has been performed for experimental simulations of the investigation of electron and ion profiles with a LaB6DC plasma gun for plasma diagnostics and determining the characteristics of magnetized plasma focused on fusion devices [24,25].Recently,we have upgraded DiPS to improve the similarity of the plasma condition with a scrape-off layer(SOL)region in fusion devices and performed experiments on the analysis of diffusion phenomena in presheaths as studies on the fusion plasma edge transport [26].

This paper estimates the effects of dusty plasma injection on the characteristics of magnetized plasmas,especially plasma density,and is an extension of previous work in DiPS for fusion plasma edge transport.Here,we report on experiments of intrinsic dust detached from PFCs to improve realistic environments,such as dust generation through ELMs interacting with PFCs in fusion devices.Dusty plasma via a laserinduced plasma [27,28] was generated by the interactions between a high-power laser beam and a full tungsten target.Using this technique,laser ablation with the melting and evaporation threshold of target materials was induced to produce transient plasmas with dust particles.To investigate the interaction of the background plasma with dusty plasma in terms of plasma discharge currents and magnetic flux intensity in DiPS,radial profiles of the plasma density at the interaction region were measured using a fast scanning probe(FSP)system with triple tips.The topics covered here are the experimental set-up in section 2,results in section 3,and conclusion in section 4.

2.Experimental setup

Figure 1.(a) Experimental set-up,(b) cross-section view for data acquisition system in divertor plasma simulator (DiPS) and (c)schematic view of an electric probe (tip area: 1.73 × 10-2 cm2).(1) dusty plasma,(2) tungsten target,(3) triple probe,(4) magnetic nozzle throat,(5) magnetic nozzle exit,(6) laser beam,(7) distance between laser beam and probe tip (d1 = 0.5 cm),(8) core plasma,(9)edge plasma,(10)oscilloscope,(11)computer,(12)linear position transducer,(13) DC power supply,(14) laser controller,(15) Nd:YAG laser,(16) probe signal (Isat, V1, V2),(17) distance signal(r),(18) fast scanning probe system (FSP),(19) distance between tungsten targets and electric probes (d2 = 5 cm) and (20) plasma.

Table 1.Comparison of plasma parameters at edge region in tokamak with DiPS.

Figure 2.(a) The region of experimental set-up in divertor plasma simulator (DiPS),(b) simulation results for magnetic intensity and(c) magnetic field profile in DiPS.

Figure 3.Frequency comparison: (a) edge localized modes (ELM)peaks in Korea Superconducting Tokamak Advanced Research(KSTAR) by electric probe measurements and (b) reflection light from interaction of laser beam with tungsten target measured by a photodiode.

Figure 4.(a) Results of interaction of laser beam with a tungsten target by using a fast camera and (b) a photograph of scanning electron microscope (SEM).

DiPS with a plasma gun can produce plasma parameters at edge-relevant plasma,as shown in table 1.It has a magnetic nozzle for the formation of a magnetic hill and geometrical production of a bounded presheath,which experimentally induces a more similar environment with the magnetized plasma at the SOL region in the fusion device[26].Figure 1 shows the experimental set-up used to analyze the effects of dusty plasma injection on the characteristics of the magnetized plasma using a high-power pulsed laser in DiPS.As shown in figure 2,the full tungsten targets located at the 4°–5° magnetic field line due to a tungsten monoblock at the ～3°–4°magnetic field line angle of incidence in the baseline ITER plasma,which are used in the fusion plasma field [15,29,30],have the following specifications:size = 25 × 25 mm2,density = 19.20 g ccm-1,hardness = 453 HV30,and tungsten content = 99.95%.An Nd:YAG pulsed laser is used for the laser ablation of the tungsten targets,with the following specifications: wavelength = 532 nm,pulse width ～10 ns,maximum energy =250 mJ,and beam diameter = 5 mm.

Since the generated dusty plasma strongly depended on the absorbed laser energy and the melting threshold of the tungsten walls for ITER was shown as a heat flux factor of transient damage threshold＞50 MJ·m-2· s-1/2per a transient heat flux such as ELMs[13,15,31,32],the transient energy flux of the Nd:YAG pulsed laser is fixed at ～100 MJ·m-2· s-1/2per laser shot.

Figure 5.Results of radial plasma density(ne)and electron temperature(Te)by using triple probes(TP).(a)ne,(b)Te at different discharge currents(10–20 A)and the fixed magnetic flux density 1 kG,(c)ne and (d)Te at different magnetic flux densities (0.8–1 kG)and the fixed discharge current 20 A.0 mm at x-axis represents plasma center.(P) and (D) in the legends are for pure plasmas and pure plasmas + injection of dusty plasmas,respectively.

These quasi-periodic bursts in fusion devices occur at a frequency of about 10–200 Hz,called‘type-I ELMs’.As shown in figure 3,beam injection frequency is fixed at 20 Hz.The distance (d) between the plasma center and tungsten target is 5 cm,which is in a presheath formed due to geometry of magnetic nozzle at DiPS,as shown in figure 1.From the measurement of the weight difference of the tungsten targets before and after laser irradiation,analysis of scanning electron microscope (SEM) photographs and a fast camera,generated dusts with a generation rate ～3 μg s-1and size of 1–10 μm were observed as shown in figure 4.

The detailed specification of DiPS and the data collection system by using an FSP system with various probe tips,including the geometry,magnetic flux intensity,plasma parameters at a steady state condition,and plasma gun,are described by Chung and Kang [23,26].In this experiment,the base pressure was 3 × 10-6Torr (4 × 10-4Pa) and operating pressure was ～7 × 10-3Torr (0.9 Pa) for argon gas at 90–130 sccm.The LaB6heating current and bias were 280 A and 16 V,respectively.The magnetic flux density was changed to 0.8–1 kG at the experiment region.Operating ranges of plasma discharge currents were 10–20 A at 40–50 V of DC bias.The incident angle of dusty plasma injection to background plasma was fixed at 90°.The input rates and size of dust particles were assumed from the measured results(3 μg s-1and 1–10 μm).

To investigate the effects of dusty plasma injection on the characteristics of background plasmas in terms of plasma discharge currents (Idis.) and magnetic flux intensity (B) at DiPS,an FSP system with triple probe tips was used for the measurement of radial plasma profiles,which are able to scan the radial plasma profiles with scan speed 1 m s-1.Radial distance was converted from bias signals measured by using linear position transducer.The structure of a triple probe,which consisted of three molybdenum tips and a ceramic insulator between the tips,is shown in figure 1(c).

3.Results

Figure 6.Results for plasma density near core region (0 ＜ r ＜25 mm).In the legends,Np and Nd are plasma densities of pure plasmas and pure plasmas with an injection of dusty plasma,respectively.

Figure 7.Results for ratio(εn)of density perturbation by injection of dusty plasma according to radial distance.(a) At different discharge currents (10–20 A) and (b) at different magnetic flux density(0.8–1 kG).0 mm at x-axis represents plasma center.

Figure 5 shows the results of measurements for the radial plasma profiles.As for the triple probe system,a fixed bias voltage -100 V was applied to the two probe tips (V1),and the other probe was for measuring the floating potential(V2).By using simple formula with potential difference,the electron temperature (Te) and plasma density (ne) were calculated usingTe= [e(V1-V2)]/ ln 2 = [e(V1-V2)]/0.693 and,where e,α,As,and k are electron charge,coefficient for the collective ion saturation current,sheath area,and the Boltzmann constant,respectively.In this study,As≈ probe tip area is assumed and α = 0.49 is used for B ＞ 0 [33].For investigation of the magnetized plasma in DiPS,probe circuits,FSP performance,raw data,and the analysis process of triple probe measurements are reported in previous work [26].We follow the same procedures as previous works for analysis of neand Te.As shown in figure 5,the general tendencies of normal magnetized plasmas are observed.For example,the increase of neand Teis found when increasing Idisand dispersion of plasma profiles is shown with the reduction of B.For focusing effect of plasma density perturbation,nedata near plasma core region was analyzed,where plasma core and edge are defined as 0 ＜ r ＜ 25 mm and r ＞ 25 mm,respectively,due to geometry of magnetic nozzle at DiPS.The average value for five data of plasma density near core region is shown in figure 6,where Npand Ndare plasma densities of pure plasmas and pure plasmas with an injection of dusty plasma,respectively.To obtain the numerical estimation of the density perturbation from dusty plasma injection by laser irradiation on a tungsten target,as shown in figure 7,εn(%) = ∣(N0-NE) /N0∣×100 according to radial density profiles was calculated,where N0and NEare plasma densities of pure plasmas and pure plasmas with the ELM-like condition,respectively,which is an injection of dusty plasma.εn～ 10%is found in core plasmas.Perturbation of neby dusty plasmas is slightly increased from core to edge plasmas,which increases by ～10%–20%.The maximum value ofεnwith ＞80% is observed in a region of the edge plasma with B = 0.9 kG and Idis= 20 A.In this study,the effects of dusty plasma injection (dusts with a generation rate of ～3 μg s-1and size of 1–10 μm)on plasma density in the magnetized plasma with plasma parameters(ne= (1–5) × 1011cm-3and Te= 10–20 eV scales at the core region in the steady state condition)are insignificant with 10% uncertainty.However,dust particles are intensively bombarded by electrons,ions,and neutral atoms in various charge states and they affect melting and sublimation temperatures,causing the phase change in matter and enhanced evaporation to produce fluctuations of neand Tein plasmas [4].

Figure 8.Result of ion saturation currents (Isat)measured by a fixed probe at core plasmas (r = 0) with B = 1 kG and Idis = 4.5 A.Black square is raw data and red line is for exponential fitting.λτ is the decay length of ion saturation currents in time scales for fluctuation duration.

Figure 9.Results of (a) ψ and (b) λτ according to ratio of plasma densities.ψ = Id/Ip is normalized factor of density fluctuations where Ip and Id are Isat values of pure plasmas and pure plasmas + injection of dusty plasmas,respectively.In figure 9(a),red line(L)is linear fitting at ranges Idis = 1–5 A.λτ is decay length of ion saturation currents in time scales for fluctuation duration.n0 is 2 × 1011 cm-3 at Idis = 10 A.

The effects of dusty plasma injection,generated from evaporation,erosion,and mass loss of a tungsten target after the interaction of high energy flux with the tungsten target,show lower εn,although a maximum εn＞ 80% was estimated.Why did dusty plasma effects result in an insignificant difference to steady state plasmas? We have the following assumptions: (i) the background plasma density is relatively higher than dusty plasma,and (ii) the duration of dust remaining in the magnetized plasma is too short to affect the plasma parameters during transient phenomena.To check the assumptions of this experiment,transient phenomena were investigated with different configurations of the experimental set-up,such as changing the scanning probe to a fixed probe and changing from high density plasmas (Idis= 10–20 A) to low density plasmas (Idis= 1–5 A).To avoid damage to probes such as melting the tips,we could not measure plasma parameters with the condition of Idis＞ 5 A since melting of a fixed probe was found in an experimental test of a fixed probe with high-density plasmas and estimation for the duration in which the probe remains in the unmelted state of the probe[34].Figure 8 shows the results of Isatmeasured by a fixed probe at core plasmas (r = 0) with B = 1 kG and Idis= 4.5 A.The transient effect of dusty plasma from laser irradiation on a tungsten target on Isat,which is a mightily important parameter for ne,in the magnetized plasmas with ne～1 × 1011cm-3is observed,as shown in figure 8.The normalized factor for fluctuation of the ion saturation current,ψ=Id/Ip,where Idand Ipare ion saturation currents of pure plasma with/without injection of dusty plasma,respectively,is introduced to estimate transient perturbation of collective ion saturation currents.Results of ψ according to the ratio of plasma densities by changing Idisare given in figure 9(a).The red line is linear fitting at ranges Idis= 1–5 A for low density plasmas.In this setup nxis plasma density at Idis= 1–5 A(n0:2 × 1011cm-3at Idis= 10 A).Perturbations of Isat,which has a higher dependence on plasma density,are found at low density plasmas.From this result with an extrapolation method,ψ = 1–2 is reached at 0.5 ＜ nx/n0＜ 0.6 for εn～ 10%.From figure 8,the duration of density perturbation by the dusty plasma is maintained for a short time of ～0.4 ms at B = 1 kG and Idis= 4.5 A.The exponential decay of density due to collisions with the mean free path could be expressed asn(t) ≈n0exp[-t/λτ]whereλτis the decay length of ion saturation currents in time scales.To estimate a duration time of perturbation,λτaccording to the ratio of plasma density was analyzed.The range from a high(peak)to low levels of Isatfor exponential fitting was used,as shown in figure 8.The tendency ofλτ,which decreases with increasing neand seems to saturate nx/n0＞ 0.35,is found as shown in figure 9(b).It is shown that the duration time of perturbation is highly dependent on plasma density [35] and has a shorter time at a higher plasma density (＞1011cm-3)than ～0.1 ms.These results presented undesirable effects from vaporization,erosion,melting,and cracks of tungsten surfaces on the plasma density stability of the edge region (ne=1012–1013cm-3) will be insignificant if the interaction of the transient high heat flux,such as the ELM-like condition with～100 MJ · m-2· s-1/2and 20 Hz,with PFCs occurs at one region in fusion devices.

4.Conclusion

The effects of dusty plasma injection on the characteristics of plasma density in magnetized plasmas were experimentally estimated at steady state and transient conditions when a similar environment for interaction of ELMs on first walls was experimentally simulated by using the interaction of a high energy pulsed laser with tungsten targets at magnetized plasmas in DiPS.For the estimation of the effects of dusty plasma injection to edge plasmas,when the interaction of ELMs ～ 100 MJ · m-2· s-1/2with 20 Hz on first walls or a divertor in tokamaks is assumed,one wave of plasma parameters for ～0.1 ms can be produced,and it consistently results in perturbation with εn～ 10% at the core region of steady state plasmas.As neincreases,the scale of the effects of dusty plasma injection on edge-relevant plasma was decreased and the duration of transient fluctuation by dusty plasma was reduced.In other words,the effects of dusty plasma injection on the plasma density of edge-relevant plasma were highly dependent on neof the edge-relevant plasmas when the ELM-like condition was replicated by exposing a pulsed laser beam to a tungsten surface.However,additional simulation and experiment studies are essential to verify the results in complex plasma with higher dust generation rate.This is because ELM phenomena with various frequencies and energies occur simultaneously and in many regions due to significantly changing plasma conditions,thus leading to increased undesirable effects on the plasma density stability from higher dust or impurity production.

Acknowledgments

This research was supported by National R&D Program through the Nation Research Foundation of Korea (NRF) funded by the Ministry of Education (2017R1D1A1B03033076).Additionally,this research was supported by National R&D Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Science,ICT & Future Planning (2019M1A7A1A03088471).